Proyectos

En este proyecto, se busca general nuevas tecnologías que permitan mejorar el manejo de recursos hídricos en la sexta región.

En este proyecto, se busca general nuevas tecnologías que permitan mejorar el manejo de recursos hídricos en la sexta región.

En este proyecto, se busca general nuevas tecnologías que permitan mejorar el manejo de recursos hídricos en la sexta región.

En este proyecto, se busca general nuevas tecnologías que permitan mejorar el manejo de recursos hídricos en la sexta región.

En este proyecto, se busca general nuevas tecnologías que permitan mejorar el manejo de recursos hídricos en la sexta región.

El aumento de los recursos tecnológicos y la automatización de los procesos agrícolas ha contribuido en gran medida al desarrollo de nuevas tecnologías que permiten caracterizar y evaluar el fenotipo de las plantas. Específicamente, las tecnologías de fenotipado de plantas son muy importantes para acelerar los programas de mejoramiento en cultivos agronómicamente importantes, y contribuyen en el proceso de selección para el desarrollo y posterior lanzamiento al mercado de nuevas variedades y cultivares. Los sistemas de visión unidimensionales (1D) y bidimensionales (2D) han sido una parte integral de la implementación exitosa de la automatización agrícola y la robótica en los procesos agrícolas. Sin embargo, las técnicas basadas en imágenes en 2D son insuficientes para investigar las estructuras espaciales de las plantas. En este sentido, la reconstrucción por medio de imágenes 3D de plantas y la adquisición de su información espacial es una forma alternativa eficaz para resolver estos problemas. En este sentido, el objetivo del presente proyecto es obtener de forma automática la estructura tridimensional del sistema de arquitectura de raíces de una planta e identificar parámetros morfológicos asociados con dicha estructura. Para ello se propone construir un prototipo, a partir de cámaras de bajo costo, para la reconstrucción automatizada de la estructura 3D de plantas, junto con la posterior detección y medición de los rasgos morfológicos de esta misma.

Introduction: Ultrasound (US) exams are extensively used in Chile and around the world. This non-invasive imaging technique has many advantages when compared to magnetic resonance, computed tomography, and others because it has a low cost, it does not need ionizing radiation, and it is portable equipment. However, this technique has many challenges; the most known is the balance between resolution and penetration depth. Recently, in 2011, a new technique of US has been described: the ultrasound localization microscopy (ULM). However, it was only in 2015 that this technique gained knowledge with the publication of Errico et al. (2015) who described the ultrafast ultrasound localization microscopy applied in vivo in rats’ brain. ULM eliminates the challenge of the balance between resolution and penetration; but a new challenge emerges: the balance between localization precision of microbubble, microbubble concentration, and acquisition time. The microbubbles (MB) are the contrast agent for the US technique. They have 1-5 µm in diameter and act as a blinking source. These MB are injected into the bloodstream and flow into the circulatory system. ULM is also known as super-resolution imaging; it can produce vascular images with a resolution around 10 µm, 10 times better than the conventional US image. This unprecedented resolution has numerous potential applications. In particular, ULM would have a high impact in oncology because the vascular structure of early tumors, that are in the range of 5 µm to 80 µm, provides information that can help in the early diagnosis and monitor therapy responses. The huge potentiality of ULM has produced a lot of excitement and expectation worldwide, and it became a hot topic in the medical ultrasound community. Unfortunately, this technique is not yet clinically approved because it is still in development stages and presents many challenges that must be solved before translating it into clinics. The mainly limitations to overcome before translating ULM into clinics are the following: contrast-to-tissue ratio (CTR), signal-to-noise ratio (SNR), acquisition time, microbubble concentration, motion, lack of a gold standard, data overdose, exploitation of ultrafast scanner uncommon in the clinic and so on. Therefore, the aim of this study is to optimize the technique to localize microbubbles, and to explore the physics of microbubble to provide a change in the paradigm of the processes to produce ULM by combining the superresolution processing with a controlled exterior force impulse. To achieve this aim, first a numerical study will be performed to simulate microbubbles into small vessels and find a better way to localize them. The robustness of the algorithm will be increased to consider the non-linear interactions between MB and US and to consider the parabolic velocity profile of the vessel/tube. Up to date, the studies on microbubbles localization in ULM are after the image acquisition. There are no tissue/flow simulations of the behavior of microbubbles into the vessels with this technique. To perform the simulation, we will use a computer with excellent storage capacity and great velocity of processing together with the MATLAB algorithm: The Full Wave Solver. Second, an experimental study will be performed to generate super resolution images in a conventional phantom. The state of art will be applied with a phantom made by microtubes and microbubbles and then, the improvement of the aim 1 will be considered into this experiment. The complexity of the phantom will be increased from a medium with only water, then gelatine and finally, we will add some respiratory simulated movement. To the experimental setup we will use the Verasonics Vantage 128 research ultrasound scanner with different types of ultrasound transducers, microbubbles, microtubes, gelatine and the respiratory movement will be simulated with a vibration testing system (Shaker VTS-100). Finally, the physics of microbubble will be explored to provide a change in the paradigm of the processes to produce ULM and to detect the MB in a more direct way, without the need to perform a filter, like the signal value decomposition (SVD). We want to apply the an external know push pulse that will produce differences in the shear waves between the microbubbles and the tissue around and with simulations we will be able to know the response of microbubbles and it may help us to separate them from the tissue. Expected results: As result of this study, we expect to develop a numerical simulation to the ULM method, by considering interactions of the US with the tissue and fluid dynamics of the blood into the vessel and significantly optimize the techniques of MB detection. Besides that, this project will help to improve the first fully programmable ultrasound scanner system in Chile. Potentially, this would open new research areas at the country level, such as ultrasound imaging, ultrasound super-resolution imaging and soft tissue characterization.